Effectiveness of Platinum-Based Chemotherapy after Progression on Poly ADP Ribose Polymerase (PARP) Inhibitor in Epithelial Ovarian Cancer (042)

Abstract

<b>Objectives:</b> PARP inhibitors have revolutionized ovarian cancer care with dramatic improvements in progression-free (PFS) and overall survival when used as maintenance therapy following platinum-based treatment. However, much remains unknown about the response to subsequent treatment after progression on PARP inhibitor (PARPi) therapy, including whether a 6-month platinum-free interval is an appropriate benchmark for using a platinum-based regimen. Therefore, we aimed to assess response to platinum-based chemotherapy in women with platinum-sensitive recurrent epithelial ovarian cancer (EOC) after progression on PARPi. <b>Methods:</b> We performed a retrospective chart review of all cases of EOC prescribed a PARPi at a tertiary care center from 2015 to 2020. We included patients who progressed on PARPi and were subsequently treated with platinum-based chemotherapy. Patients who received fewer than six months of PARPi treatment were excluded. The primary outcome was PFS on the platinum regimen (ptPFS), defined as the time from the start of platinum therapy to physician documented progression. Germline and somatic <i>BRCA</i> testing, number of prior lines of chemotherapy, duration of therapy with PARPi and subsequent platinum regimen, and demographic data were collected for each case. An accelerated failure time model was developed to assess the primary outcome with a multivariable model including <i>BRCA</i> status, prior lines of chemotherapy, and time on PARPi in 6-month intervals. <b>Results:</b> A total of 283 charts were reviewed, and 40 cases met the inclusion criteria. Of these, 60% (<i>n</i>=24) had a <i>BRCA</i> mutation (<i>BRCA</i>mut); 75% (<i>n</i>=18) were germline and 25% (<i>n</i>=6) were somatic mutations. Twenty percent (<i>n</i>=8) of patients received a PARPi after first-line treatment, 45% (<i>n</i>=18) after second, 28% (<i>n</i>=11) after third, and 7% (<i>n</i>=3) after fourth. Sixty percent (<i>n</i>=24) of patients were prescribed olaparib, 20% (<i>n</i>=8) niraparib, 10% (<i>n</i>=4) rucaparib, and 10% received two agents due to toxicity with their first. The median time on PARPi was 411 days, and the median ptPFS was 218 days. In the multivariable model (Table 1), <i>BRCA</i> mutations were associated with decreased ptPFS (<i>BRCA</i>wt 274 days vs <i>BRCA</i>mut 170 days, p=0.004). The median time on PARPi greater than 18 months was associated with improved ptPFS compared to 12-18 months and <12 months (262 vs 188 vs 163 days, respectively, p=0.007). The number of prior lines of therapy was not associated with ptPFS (p=0.13). <b>Conclusions:</b> This retrospective analysis supports the ongoing use of platinum-based chemotherapy in recurrent EOC after progression on PARPi therapy with outcomes comparable to those seen in OCEANS and GOG-213. These data suggest that resistance to PARPi does not necessarily imply platinum resistance. Poorer ptPFS in tumors with a <i>BRCA</i>mut compared to those without was unexpected. While the data surrounding post PARPi treatment response is limited, growing evidence from the recently reported OReO trial corroborates this finding. This raises the question of a difference in the underlying mechanism leading to platinum resistance and PARPi resistance in <i>BRCA</i>mut versus <i>BRCA</i>wt EOC. Molecular assessment of recurrent tumors with clinical correlation will be important to optimize treatment approaches.